Re: [AMBER] X-OS-P-X and X-os-p5-X AMBER99SB and GAFF parameters mismatch

From: Alan <>
Date: Thu, 14 Jan 2016 09:13:55 +0000

Dear Tarak, ACPYPE should work fine with GMX 4.5 (old GMX 4.0 had only RB

Check if you using the last version of ACPYPE, few changed regarded default

Pay attention to these options:

  -a ATOM_TYPE, --atom_type=ATOM_TYPE
                        atom type, can be gaff or amber (AMBER14SB), default
                        is gaff

  -z, --gmx4 write RB dihedrals old GMX 4.0 (default is NOT, for
GMX >=4.5)

(whatever option you take whether using -z or not, GMX will pick the right

And be aware that you ligand may not have the right charge distribution
profile, so it may run fine alone, but when in the ligand pocket site of
the protein, high repulsive forces may show up.

| ACPYPE: AnteChamber PYthon Parser interfacE v. 2015-11-02 10:03:43Z Rev:
8810 (c) 2016 AWSdS |
    acpype -i _file_ [-c _string_] [-n _int_] [-m _int_] [-a _string_] [-f]
etc. or
    acpype -p _prmtop_ -x _inpcrd_ [-d]

    output: assuming 'root' is the basename of either the top input file,
            the 3-letter residue name or user defined (-b option)
    root_bcc_gaff.mol2: final mol2 file with 'bcc' charges and 'gaff' atom
    root_AC.inpcrd : coord file for AMBER
    root_AC.prmtop : topology and parameter file for AMBER
    root_AC.lib : residue library file for AMBER
    root_AC.frcmod : modified force field parameters
    root_GMX.gro : coord file for GROMACS : topology file for GROMACS
    root_GMX.itp : molecule unit topology and parameter file for
    root_GMX_OPLS.itp : OPLS/AA mol unit topol & par file for GROMACS
    em.mdp, md.mdp : run parameters file for GROMACS
    root_NEW.pdb : final pdb file generated by ACPYPE : topology file for CNS/XPLOR
    root_CNS.par : parameter file for CNS/XPLOR
    root_CNS.inp : run parameters file for CNS/XPLOR
    root_CHARMM.rtf : topology file for CHARMM
    root_CHARMM.prm : parameter file for CHARMM
    root_CHARMM.inp : run parameters file for CHARMM

  -h, --help show this help message and exit
  -i INPUT, --input=INPUT
                        input file name with either extension '.pdb', '.mdl'
                        or '.mol2' (mandatory if -p and -x not set)
  -b BASENAME, --basename=BASENAME
                        a basename for the project (folder and output files)
  -x INPCRD, --inpcrd=INPCRD
                        amber inpcrd file name (always used with -p)
  -p PRMTOP, --prmtop=PRMTOP
                        amber prmtop file name (always used with -x)
  -c CHARGE_METHOD, --charge_method=CHARGE_METHOD
                        charge method: gas, bcc (default), user (user's
                        charges in mol2 file)
  -n NET_CHARGE, --net_charge=NET_CHARGE
                        net molecular charge (int), for gas default is 0
                        multiplicity (2S+1), default is 1
  -a ATOM_TYPE, --atom_type=ATOM_TYPE
                        atom type, can be gaff or amber (AMBER14SB), default
                        is gaff
  -q QPROG, --qprog=QPROG
                        am1-bcc flag, sqm (default), divcon, mopac
  -k KEYWORD, --keyword=KEYWORD
                        mopac or sqm keyword, inside quotes
  -f, --force force topologies recalculation anew
  -d, --debug for debugging purposes, keep any temporary file
  -o OUTTOP, --outtop=OUTTOP
                        output topologies: all (default), gmx, cns or charmm
  -z, --gmx4 write RB dihedrals old GMX 4.0
  -t, --cnstop write CNS topology with allhdg-like parameters
  -e ENGINE, --engine=ENGINE
                        engine: tleap (default) or sleap (not fully matured)
  -s MAX_TIME, --max_time=MAX_TIME
                        max time (in sec) tolerance for sqm/mopac, default
                        10 hours
  -y, --ipython start iPython interpreter
  -w, --verboseless print nothing
  -g, --disambiguate disambiguate lower and uppercase atomtypes in GMX
  -u, --direct for 'amb2gmx' mode, does a direct conversion, for
  -l, --sorted sort atoms for GMX ordering
  -j, --chiral create improper dihedral parameters for chiral atoms
                        in CNS

On 13 January 2016 at 13:42, tarak karmakar <> wrote:

> Dear All,
> I have a confusion related to two dihedral parameters that are present in
> ffbonded.itp file,
> X OH P X 9 0.0 1.04600 3 ; JCC,7,(1986),230
> X OS P X 9 0.0 1.04600 3 ; JCC,7,(1986),230
> the similar dihedral parameters from parm99SBgaff.dat (GAFF) file,
> X -oh-p5-X 3 1.600 0.000 3.000
> X -os-p5-X 3 2.400 0.000 2.000
> As you see, there is a mismatch between the periodicity (last column) of
> the similar dihedral angle from AMBER99SB and gaff databases.
> Would anybody like to comment on this?
> I just read an earlier discussion in the following link,
> But, my problem is little more than the above.
> I am currently simulating a protonated pyrophosphate ligand molecule in a
> protein. During the heating step in NVT ensemble (within 150-200 ps, 200 K
> temp), the atoms present in the phosphate groups (P,O and H) are showing
> bizarre bonds and angles. I tried with the higher force constants (10 times
> than the existing one) for the above mentioned dihedral angles, and then
> the molecule was behaving fine.
> Has anybody experienced this type of issue?
> I appreciate any suggestions.
> Regards,
> Tarak
> _______________________________________________
> AMBER mailing list

Alan Wilter SOUSA da SILVA, DSc
Senior Bioinformatician, UniProt
European Bioinformatics Institute (EMBL-EBI)
European Molecular Biology Laboratory
Wellcome Trust Genome Campus
Cambridge CB10 1SD
United Kingdom
Tel: +44 (0)1223 494588
AMBER mailing list
Received on Thu Jan 14 2016 - 01:30:02 PST
Custom Search